Epithelial ovarian cancer (EOC) is the fifth leading cause of cancer mortality in women in the US but little is known of its etiology. Mitochondria participate in a number of essential functions in the cancer cell. The mitochondrial genome consists of a circular mitochondrial DNA (mtDNA) of 16.5 kilobases and approximately 1500 nuclear mitochondrial genes. Germline and somatic mitochondrial genome variants and mutations have been identified in a number of different cancers. However, very few studies have investigated the mitochondrial genome in EOC. We have performed genome-wide association studies (GWAS) and found a significant number of single nucleotide polymorphisms (SNPs) associations at the significance level (p<10-4 but >10-8) in genes showing significant associations in oxidative phosphorylation, epithelial-mesenchymal transition and apoptosis pathways, where the mitochondria plays a key role. The common element of the central role of mitochondria in the three pathways showing significant associations provides evidence of cross-talk between the mitochondrial and nuclear genomes, which warrants further investigation. We previously identified associations with ovarian cancer risk in apoptosis-related and EMT-related genes. Taken together, these results prompted this study, the GOAL of which is to more comprehensively investigate the contribution of mitochondrial genome variation to ovarian cancer risk. Nuclear mitochondrial gene variation, mtDNA haplotype analysis and correlative somatic mtDNA studies will be performed. These studies will enable us to fully evaluate the HYPOTHESIS that inherited variation in the mitochondrial genome as well as the nuclear genome should be considered as factors for risk of epithelial ovarian cancer. The specific aims are: 1) Stage 1 - To investigate SNPs in nuclear mitochondrial-related and mtDNA genes. 2) Stage 2 - To validate the top hits in nuclear mitochondrial-related genes and mtDNA genes in an independent case-control collection 3) To identify mtDNA mutations and variants in EOC. This study is novel as we are investigating both nuclear and mitochondrial variation as risk factors for ovarian cancer. There are few current studies on mitochondria and ovarian cancer. The results of this study will contribute significantly to our broader efforts to elucidate the genetic basis of ovarian cancer and unravel the complexities of this lethal disease. PUBLIC HEALTH RELEVANCE: Epithelial ovarian cancer (EOC) is the fifth leading cause of cancer mortality in women in the US but little is known of the genetic basis of this disease. Mitochondria participate in a number of essential functions in the cell including energy production, protein and amino acid synthesis and programmed cell death (apoptosis). The mitochondrial genome consists of a circular mitochondrial DNA (mtDNA) of 16.5 kilobases which codes for 37 genes. There are also approximately 1500 mitochondrial genes in the nuclear DNA, involved in a myriad of essential cellular functions such as signaling, cell growth and differentiation and apoptosis. Germline and somatic mitochondrial genome variants and mutations have been identified in a number of different cancers. However, very few studies have investigated the mitochondrial genome in EOC. We have previously performed a study looking at genetic variants around the whole genome in women with ovarian cancer compared to those who do not have the disease. Some of the most significant associations with ovarian cancer are in mitochondria-related genes. Furthermore we have previously looked at some genes related to the mitochondria and found some associations with ovarian cancer. The GOAL of the proposed study is to more comprehensively investigate mitochondrial genome variation in ovarian cancer. We will look both in the germline and tumor DNA. Nuclear mitochondrial gene variation, mtDNA haplotype analysis and correlative somatic mtDNA studies will be performed. These studies will enable us to fully evaluate the HYPOTHESIS that inherited variation in the mitochondrial genome as well as the nuclear genome should be considered as factors for risk of epithelial ovarian cancer. We have the following specific aims: 1) Stage 1 - To investigate SNPs in nuclear mitochondrial-related and mtDNA genes Approach: We will look at the results of the nuclear mitochondrial-related and mtDNA genes from the oxidative phosphorylation, epithelial-mesenchymal transition, protein synthesis and apoptosis pathways. Then for those genes that are significantly associated with ovarian cancer we will expand the coverage on genes in our chosen pathways in a study population of 2000 EOC cases and 2000 controls. Additional mtDNA SNPs will be included to ensure identification of the most common mtDNA haplogroups. We will use statistical analysis to look for SNP associations within major histologic subtypes and tumor characteristics. 2) Stage 2 - To validate the top hits in nuclear mitochondrial-related genes and mtDNA genes in an independent case-control collection Approach: We will validate the SNPs showing the most statistically significant associations with ovarian cancer (approximately 40 SNPs) in a collection of up to 10000 ovarian cancer cases and 10000 frequency matched controls from the Ovarian Cancer Association Consortium (OCAC). Furthermore we will include mtDNA SNPs that define the most common mtDNA haplogroups. We will then use logistic regression to investigate the association of mtDNA haplogroups and histologic subtype and perform exploratory gene-environment interactions with non-genetic risk factors. 3) To identify mtDNA mutations and variants in EOC Approach: We will undertake mtDNA sequencing in ovarian tumor tissue in a subset of 400 ovarian cancer cases from the above panel. Sequencing will be performed using the Affymetrix MitoChip v2.0. Variants and mutations will be investigated in germline DNA of the same individual . We will investigate if particular mtDNA haplotypes are correlated with somatic variants or mutations in ovarian tumor tissue or particular variants are associated with tumor-specific characteristics. In this unique and powerful study we will investigate the impact of variation in the mitochondrial genome and variation of the nuclear genome on ovarian cancer risk. The preliminary data for this study comes from two ovarian cancer GWASes in over 8000 individuals which provides an immense resource upon which to build this study so as to make strides in ovarian cancer genetic research. The epidemiologic and genotype datasets are readily available, and are currently housed at the Moffitt Cancer Center which proves the feasibility of the proposed research. The blood DNAs and tumor specimens/blocks are also readily available for this study. The proposed work is scientifically sound, highly significant and novel. Our research team is highly skilled and has been working together successfully for many years in ovarian cancer SNP association studies as is evidenced by our joint publications and funding record. We will address several priorities established by the National Cancer Institute-convened Gynecologic Cancers Progress Review Group including (1) application of novel study designs to elucidate the genomics of ovarian cancer, and (2) systematic screening of germline genetic targets with translational impact. Finally, this study will address a number of the questions on the role of mitochondria in cancer that were posed in the program announcement (PA-08-143).